Abstract
Background: As crime is increasing day by day, various applications are proposed to protect public places. Monitoring and tracking of video surveillance system is the most difficult task and it is prominent that human beings are not reliable and efficacious in doing this job.
Objective: The prime objective of this research is to develop an automatic monitoring and inspecting system that is competent enough to detect and track the moving objects in real-time using a low-resolution video surveillance camera.
Methods: Firstly, the video data acquired from a low-resolution video surveillance camera is used for generating RGB video frames which are converted into gray scale. Optical flow and Eigen face algorithms are applied to extract and match the moving object in the video sequence with the images stored in the database.
Results: The proposed system is compared with the already existing systems and it is observed that this approach gives more accurate results. This system can meet the requirement of real-time tracking even when the targeted image resolution is smaller than 160x120.
Conclusion: This method uses optical flow and Eigen face algorithm to track and detect the moving objects. The system gives high performance and can be used for real time object tracking. The same experiment can be applied for the human faces too.
Keywords: Eigen face algorithm, optical flow, background subtraction, video processing, image processing, pattern recognition.
Graphical Abstract